Training in progress, step 3500, checkpoint
Browse files
last-checkpoint/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 9446744
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8e72f74f4e0e6b9d2ea75ad255fa3eaeba01a7e0823bcb5a98cb4e80510a589e
|
| 3 |
size 9446744
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4879947
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:200b11c2ef8037da35c5780017b6651f6b6076fc914de77cf6066270f2e1b523
|
| 3 |
size 4879947
|
last-checkpoint/rng_state_0.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14917
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f531f0a7c41eca7b2e7efd3f8d7ba2d9d38c29a18b748522bac1a5c64a9df8e8
|
| 3 |
size 14917
|
last-checkpoint/rng_state_1.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14917
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4ef97fc93f1e58ebc9e20759b520e3bc8c347e36f8f62fa0c9adef310bbf48e6
|
| 3 |
size 14917
|
last-checkpoint/scaler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1383
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ac1b9358ab1cb9acff90f4d1d692ac08bbdb1986a66544aed9f6fe9b801b17b3
|
| 3 |
size 1383
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1465
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f2198d94a2d44839615c5c39451a8a9843983d5979dc85a343072f765e711e24
|
| 3 |
size 1465
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -2,9 +2,9 @@
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
-
"epoch":
|
| 6 |
"eval_steps": 500,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
@@ -3008,6 +3008,506 @@
|
|
| 3008 |
"mean_token_accuracy": 0.6704145818948746,
|
| 3009 |
"num_tokens": 17831493.0,
|
| 3010 |
"step": 3000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3011 |
}
|
| 3012 |
],
|
| 3013 |
"logging_steps": 10,
|
|
@@ -3027,7 +3527,7 @@
|
|
| 3027 |
"attributes": {}
|
| 3028 |
}
|
| 3029 |
},
|
| 3030 |
-
"total_flos": 1.
|
| 3031 |
"train_batch_size": 2,
|
| 3032 |
"trial_name": null,
|
| 3033 |
"trial_params": null
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 6.013754566946056,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 3500,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 3008 |
"mean_token_accuracy": 0.6704145818948746,
|
| 3009 |
"num_tokens": 17831493.0,
|
| 3010 |
"step": 3000
|
| 3011 |
+
},
|
| 3012 |
+
{
|
| 3013 |
+
"entropy": 1.7866925299167633,
|
| 3014 |
+
"epoch": 5.171932086825704,
|
| 3015 |
+
"grad_norm": 0.775793731212616,
|
| 3016 |
+
"learning_rate": 7.964e-05,
|
| 3017 |
+
"loss": 1.8513370513916017,
|
| 3018 |
+
"mean_token_accuracy": 0.6593568369746208,
|
| 3019 |
+
"num_tokens": 17893338.0,
|
| 3020 |
+
"step": 3010
|
| 3021 |
+
},
|
| 3022 |
+
{
|
| 3023 |
+
"entropy": 1.720614206790924,
|
| 3024 |
+
"epoch": 5.1891252955082745,
|
| 3025 |
+
"grad_norm": 0.7855071425437927,
|
| 3026 |
+
"learning_rate": 7.924000000000001e-05,
|
| 3027 |
+
"loss": 1.7529998779296876,
|
| 3028 |
+
"mean_token_accuracy": 0.6738685265183448,
|
| 3029 |
+
"num_tokens": 17949102.0,
|
| 3030 |
+
"step": 3020
|
| 3031 |
+
},
|
| 3032 |
+
{
|
| 3033 |
+
"entropy": 1.8130397230386734,
|
| 3034 |
+
"epoch": 5.206318504190844,
|
| 3035 |
+
"grad_norm": 0.7261347770690918,
|
| 3036 |
+
"learning_rate": 7.884e-05,
|
| 3037 |
+
"loss": 1.8530288696289063,
|
| 3038 |
+
"mean_token_accuracy": 0.6672368694096804,
|
| 3039 |
+
"num_tokens": 18011291.0,
|
| 3040 |
+
"step": 3030
|
| 3041 |
+
},
|
| 3042 |
+
{
|
| 3043 |
+
"entropy": 1.7442916065454483,
|
| 3044 |
+
"epoch": 5.223511712873415,
|
| 3045 |
+
"grad_norm": 0.7350125908851624,
|
| 3046 |
+
"learning_rate": 7.844e-05,
|
| 3047 |
+
"loss": 1.8185455322265625,
|
| 3048 |
+
"mean_token_accuracy": 0.6648106183856726,
|
| 3049 |
+
"num_tokens": 18069969.0,
|
| 3050 |
+
"step": 3040
|
| 3051 |
+
},
|
| 3052 |
+
{
|
| 3053 |
+
"entropy": 1.8336029559373856,
|
| 3054 |
+
"epoch": 5.240704921555985,
|
| 3055 |
+
"grad_norm": 0.9380921125411987,
|
| 3056 |
+
"learning_rate": 7.804e-05,
|
| 3057 |
+
"loss": 1.8585586547851562,
|
| 3058 |
+
"mean_token_accuracy": 0.6559876747429371,
|
| 3059 |
+
"num_tokens": 18126839.0,
|
| 3060 |
+
"step": 3050
|
| 3061 |
+
},
|
| 3062 |
+
{
|
| 3063 |
+
"entropy": 1.831410789489746,
|
| 3064 |
+
"epoch": 5.257898130238556,
|
| 3065 |
+
"grad_norm": 0.7422699928283691,
|
| 3066 |
+
"learning_rate": 7.764e-05,
|
| 3067 |
+
"loss": 1.8669567108154297,
|
| 3068 |
+
"mean_token_accuracy": 0.660079612582922,
|
| 3069 |
+
"num_tokens": 18189169.0,
|
| 3070 |
+
"step": 3060
|
| 3071 |
+
},
|
| 3072 |
+
{
|
| 3073 |
+
"entropy": 1.7772829428315162,
|
| 3074 |
+
"epoch": 5.2750913389211265,
|
| 3075 |
+
"grad_norm": 0.80905681848526,
|
| 3076 |
+
"learning_rate": 7.724e-05,
|
| 3077 |
+
"loss": 1.8516859054565429,
|
| 3078 |
+
"mean_token_accuracy": 0.6614492174237967,
|
| 3079 |
+
"num_tokens": 18247092.0,
|
| 3080 |
+
"step": 3070
|
| 3081 |
+
},
|
| 3082 |
+
{
|
| 3083 |
+
"entropy": 1.8644750490784645,
|
| 3084 |
+
"epoch": 5.292284547603696,
|
| 3085 |
+
"grad_norm": 0.7613252997398376,
|
| 3086 |
+
"learning_rate": 7.684e-05,
|
| 3087 |
+
"loss": 1.8771135330200195,
|
| 3088 |
+
"mean_token_accuracy": 0.6572393793612719,
|
| 3089 |
+
"num_tokens": 18305337.0,
|
| 3090 |
+
"step": 3080
|
| 3091 |
+
},
|
| 3092 |
+
{
|
| 3093 |
+
"entropy": 1.7888765200972556,
|
| 3094 |
+
"epoch": 5.309477756286267,
|
| 3095 |
+
"grad_norm": 0.8216497302055359,
|
| 3096 |
+
"learning_rate": 7.644e-05,
|
| 3097 |
+
"loss": 1.877157211303711,
|
| 3098 |
+
"mean_token_accuracy": 0.6567147132009268,
|
| 3099 |
+
"num_tokens": 18366693.0,
|
| 3100 |
+
"step": 3090
|
| 3101 |
+
},
|
| 3102 |
+
{
|
| 3103 |
+
"entropy": 1.8046741798520087,
|
| 3104 |
+
"epoch": 5.326670964968837,
|
| 3105 |
+
"grad_norm": 0.7475964426994324,
|
| 3106 |
+
"learning_rate": 7.604e-05,
|
| 3107 |
+
"loss": 1.8193256378173828,
|
| 3108 |
+
"mean_token_accuracy": 0.6620738692581654,
|
| 3109 |
+
"num_tokens": 18424592.0,
|
| 3110 |
+
"step": 3100
|
| 3111 |
+
},
|
| 3112 |
+
{
|
| 3113 |
+
"entropy": 1.7556863978505135,
|
| 3114 |
+
"epoch": 5.343864173651408,
|
| 3115 |
+
"grad_norm": 0.7376730442047119,
|
| 3116 |
+
"learning_rate": 7.564e-05,
|
| 3117 |
+
"loss": 1.8117481231689454,
|
| 3118 |
+
"mean_token_accuracy": 0.6669185206294059,
|
| 3119 |
+
"num_tokens": 18480554.0,
|
| 3120 |
+
"step": 3110
|
| 3121 |
+
},
|
| 3122 |
+
{
|
| 3123 |
+
"entropy": 1.680773164331913,
|
| 3124 |
+
"epoch": 5.361057382333978,
|
| 3125 |
+
"grad_norm": 0.8276366591453552,
|
| 3126 |
+
"learning_rate": 7.524e-05,
|
| 3127 |
+
"loss": 1.7203754425048827,
|
| 3128 |
+
"mean_token_accuracy": 0.6801572386175394,
|
| 3129 |
+
"num_tokens": 18533859.0,
|
| 3130 |
+
"step": 3120
|
| 3131 |
+
},
|
| 3132 |
+
{
|
| 3133 |
+
"entropy": 1.7824992030858993,
|
| 3134 |
+
"epoch": 5.378250591016548,
|
| 3135 |
+
"grad_norm": 0.7689419984817505,
|
| 3136 |
+
"learning_rate": 7.484e-05,
|
| 3137 |
+
"loss": 1.7929088592529296,
|
| 3138 |
+
"mean_token_accuracy": 0.6625824831426144,
|
| 3139 |
+
"num_tokens": 18591432.0,
|
| 3140 |
+
"step": 3130
|
| 3141 |
+
},
|
| 3142 |
+
{
|
| 3143 |
+
"entropy": 1.7943954214453697,
|
| 3144 |
+
"epoch": 5.395443799699119,
|
| 3145 |
+
"grad_norm": 0.7818305492401123,
|
| 3146 |
+
"learning_rate": 7.444e-05,
|
| 3147 |
+
"loss": 1.9027202606201172,
|
| 3148 |
+
"mean_token_accuracy": 0.6574487689882517,
|
| 3149 |
+
"num_tokens": 18653780.0,
|
| 3150 |
+
"step": 3140
|
| 3151 |
+
},
|
| 3152 |
+
{
|
| 3153 |
+
"entropy": 1.6858137652277947,
|
| 3154 |
+
"epoch": 5.412637008381689,
|
| 3155 |
+
"grad_norm": 0.7783890962600708,
|
| 3156 |
+
"learning_rate": 7.404e-05,
|
| 3157 |
+
"loss": 1.6952400207519531,
|
| 3158 |
+
"mean_token_accuracy": 0.681446236371994,
|
| 3159 |
+
"num_tokens": 18712089.0,
|
| 3160 |
+
"step": 3150
|
| 3161 |
+
},
|
| 3162 |
+
{
|
| 3163 |
+
"entropy": 1.781475655734539,
|
| 3164 |
+
"epoch": 5.42983021706426,
|
| 3165 |
+
"grad_norm": 0.8033313751220703,
|
| 3166 |
+
"learning_rate": 7.364e-05,
|
| 3167 |
+
"loss": 1.8575824737548827,
|
| 3168 |
+
"mean_token_accuracy": 0.6616954285651445,
|
| 3169 |
+
"num_tokens": 18770760.0,
|
| 3170 |
+
"step": 3160
|
| 3171 |
+
},
|
| 3172 |
+
{
|
| 3173 |
+
"entropy": 1.8044284671545028,
|
| 3174 |
+
"epoch": 5.44702342574683,
|
| 3175 |
+
"grad_norm": 0.8778691291809082,
|
| 3176 |
+
"learning_rate": 7.324000000000001e-05,
|
| 3177 |
+
"loss": 1.868129348754883,
|
| 3178 |
+
"mean_token_accuracy": 0.6609551507979632,
|
| 3179 |
+
"num_tokens": 18830053.0,
|
| 3180 |
+
"step": 3170
|
| 3181 |
+
},
|
| 3182 |
+
{
|
| 3183 |
+
"entropy": 1.7783053085207938,
|
| 3184 |
+
"epoch": 5.4642166344294,
|
| 3185 |
+
"grad_norm": 0.794116735458374,
|
| 3186 |
+
"learning_rate": 7.284000000000001e-05,
|
| 3187 |
+
"loss": 1.8013723373413086,
|
| 3188 |
+
"mean_token_accuracy": 0.6687252540141344,
|
| 3189 |
+
"num_tokens": 18889827.0,
|
| 3190 |
+
"step": 3180
|
| 3191 |
+
},
|
| 3192 |
+
{
|
| 3193 |
+
"entropy": 1.7062184020876885,
|
| 3194 |
+
"epoch": 5.481409843111971,
|
| 3195 |
+
"grad_norm": 0.7348354458808899,
|
| 3196 |
+
"learning_rate": 7.244e-05,
|
| 3197 |
+
"loss": 1.7668045043945313,
|
| 3198 |
+
"mean_token_accuracy": 0.6689080417156219,
|
| 3199 |
+
"num_tokens": 18948434.0,
|
| 3200 |
+
"step": 3190
|
| 3201 |
+
},
|
| 3202 |
+
{
|
| 3203 |
+
"entropy": 1.7791135892271996,
|
| 3204 |
+
"epoch": 5.498603051794541,
|
| 3205 |
+
"grad_norm": 0.7285153865814209,
|
| 3206 |
+
"learning_rate": 7.204000000000001e-05,
|
| 3207 |
+
"loss": 1.8126005172729491,
|
| 3208 |
+
"mean_token_accuracy": 0.6612196549773216,
|
| 3209 |
+
"num_tokens": 19009735.0,
|
| 3210 |
+
"step": 3200
|
| 3211 |
+
},
|
| 3212 |
+
{
|
| 3213 |
+
"entropy": 1.7333651915192605,
|
| 3214 |
+
"epoch": 5.515796260477112,
|
| 3215 |
+
"grad_norm": 0.768817126750946,
|
| 3216 |
+
"learning_rate": 7.164000000000001e-05,
|
| 3217 |
+
"loss": 1.7593820571899415,
|
| 3218 |
+
"mean_token_accuracy": 0.6708316601812839,
|
| 3219 |
+
"num_tokens": 19070839.0,
|
| 3220 |
+
"step": 3210
|
| 3221 |
+
},
|
| 3222 |
+
{
|
| 3223 |
+
"entropy": 1.7646871596574782,
|
| 3224 |
+
"epoch": 5.532989469159682,
|
| 3225 |
+
"grad_norm": 0.7897234559059143,
|
| 3226 |
+
"learning_rate": 7.124e-05,
|
| 3227 |
+
"loss": 1.8217975616455078,
|
| 3228 |
+
"mean_token_accuracy": 0.6624716755002737,
|
| 3229 |
+
"num_tokens": 19131701.0,
|
| 3230 |
+
"step": 3220
|
| 3231 |
+
},
|
| 3232 |
+
{
|
| 3233 |
+
"entropy": 1.79796422123909,
|
| 3234 |
+
"epoch": 5.550182677842252,
|
| 3235 |
+
"grad_norm": 0.7753779292106628,
|
| 3236 |
+
"learning_rate": 7.084e-05,
|
| 3237 |
+
"loss": 1.8724674224853515,
|
| 3238 |
+
"mean_token_accuracy": 0.6603185098618269,
|
| 3239 |
+
"num_tokens": 19192206.0,
|
| 3240 |
+
"step": 3230
|
| 3241 |
+
},
|
| 3242 |
+
{
|
| 3243 |
+
"entropy": 1.6741029411554336,
|
| 3244 |
+
"epoch": 5.567375886524823,
|
| 3245 |
+
"grad_norm": 0.7559053301811218,
|
| 3246 |
+
"learning_rate": 7.044000000000001e-05,
|
| 3247 |
+
"loss": 1.7434120178222656,
|
| 3248 |
+
"mean_token_accuracy": 0.6718135714530945,
|
| 3249 |
+
"num_tokens": 19250887.0,
|
| 3250 |
+
"step": 3240
|
| 3251 |
+
},
|
| 3252 |
+
{
|
| 3253 |
+
"entropy": 1.8032452374696732,
|
| 3254 |
+
"epoch": 5.584569095207393,
|
| 3255 |
+
"grad_norm": 0.8887183666229248,
|
| 3256 |
+
"learning_rate": 7.004e-05,
|
| 3257 |
+
"loss": 1.8439495086669921,
|
| 3258 |
+
"mean_token_accuracy": 0.6600725017488003,
|
| 3259 |
+
"num_tokens": 19310066.0,
|
| 3260 |
+
"step": 3250
|
| 3261 |
+
},
|
| 3262 |
+
{
|
| 3263 |
+
"entropy": 1.7758998274803162,
|
| 3264 |
+
"epoch": 5.601762303889964,
|
| 3265 |
+
"grad_norm": 0.7295767068862915,
|
| 3266 |
+
"learning_rate": 6.964e-05,
|
| 3267 |
+
"loss": 1.8148815155029296,
|
| 3268 |
+
"mean_token_accuracy": 0.6652825616300106,
|
| 3269 |
+
"num_tokens": 19369366.0,
|
| 3270 |
+
"step": 3260
|
| 3271 |
+
},
|
| 3272 |
+
{
|
| 3273 |
+
"entropy": 1.662617878615856,
|
| 3274 |
+
"epoch": 5.6189555125725335,
|
| 3275 |
+
"grad_norm": 0.7842978835105896,
|
| 3276 |
+
"learning_rate": 6.924000000000001e-05,
|
| 3277 |
+
"loss": 1.7120464324951172,
|
| 3278 |
+
"mean_token_accuracy": 0.6784614086151123,
|
| 3279 |
+
"num_tokens": 19427377.0,
|
| 3280 |
+
"step": 3270
|
| 3281 |
+
},
|
| 3282 |
+
{
|
| 3283 |
+
"entropy": 1.8431208834052086,
|
| 3284 |
+
"epoch": 5.636148721255104,
|
| 3285 |
+
"grad_norm": 0.8514787554740906,
|
| 3286 |
+
"learning_rate": 6.884e-05,
|
| 3287 |
+
"loss": 1.9169921875,
|
| 3288 |
+
"mean_token_accuracy": 0.6530084304511548,
|
| 3289 |
+
"num_tokens": 19488230.0,
|
| 3290 |
+
"step": 3280
|
| 3291 |
+
},
|
| 3292 |
+
{
|
| 3293 |
+
"entropy": 1.6891573801636697,
|
| 3294 |
+
"epoch": 5.653341929937675,
|
| 3295 |
+
"grad_norm": 0.7638376951217651,
|
| 3296 |
+
"learning_rate": 6.844e-05,
|
| 3297 |
+
"loss": 1.7445995330810546,
|
| 3298 |
+
"mean_token_accuracy": 0.6780110366642476,
|
| 3299 |
+
"num_tokens": 19546500.0,
|
| 3300 |
+
"step": 3290
|
| 3301 |
+
},
|
| 3302 |
+
{
|
| 3303 |
+
"entropy": 1.7722659215331078,
|
| 3304 |
+
"epoch": 5.670535138620245,
|
| 3305 |
+
"grad_norm": 0.8072571754455566,
|
| 3306 |
+
"learning_rate": 6.804e-05,
|
| 3307 |
+
"loss": 1.8334453582763672,
|
| 3308 |
+
"mean_token_accuracy": 0.6662346951663494,
|
| 3309 |
+
"num_tokens": 19604586.0,
|
| 3310 |
+
"step": 3300
|
| 3311 |
+
},
|
| 3312 |
+
{
|
| 3313 |
+
"entropy": 1.824295823276043,
|
| 3314 |
+
"epoch": 5.687728347302816,
|
| 3315 |
+
"grad_norm": 0.8235921263694763,
|
| 3316 |
+
"learning_rate": 6.764000000000001e-05,
|
| 3317 |
+
"loss": 1.936505126953125,
|
| 3318 |
+
"mean_token_accuracy": 0.6559072963893413,
|
| 3319 |
+
"num_tokens": 19664352.0,
|
| 3320 |
+
"step": 3310
|
| 3321 |
+
},
|
| 3322 |
+
{
|
| 3323 |
+
"entropy": 1.776401199400425,
|
| 3324 |
+
"epoch": 5.704921555985385,
|
| 3325 |
+
"grad_norm": 0.7260850071907043,
|
| 3326 |
+
"learning_rate": 6.724e-05,
|
| 3327 |
+
"loss": 1.8124887466430664,
|
| 3328 |
+
"mean_token_accuracy": 0.6629696622490883,
|
| 3329 |
+
"num_tokens": 19725395.0,
|
| 3330 |
+
"step": 3320
|
| 3331 |
+
},
|
| 3332 |
+
{
|
| 3333 |
+
"entropy": 1.8356325037777423,
|
| 3334 |
+
"epoch": 5.722114764667956,
|
| 3335 |
+
"grad_norm": 0.7962324619293213,
|
| 3336 |
+
"learning_rate": 6.684e-05,
|
| 3337 |
+
"loss": 1.865267562866211,
|
| 3338 |
+
"mean_token_accuracy": 0.6575648851692677,
|
| 3339 |
+
"num_tokens": 19786886.0,
|
| 3340 |
+
"step": 3330
|
| 3341 |
+
},
|
| 3342 |
+
{
|
| 3343 |
+
"entropy": 1.7975003249943255,
|
| 3344 |
+
"epoch": 5.739307973350527,
|
| 3345 |
+
"grad_norm": 0.8319332599639893,
|
| 3346 |
+
"learning_rate": 6.644000000000001e-05,
|
| 3347 |
+
"loss": 1.8043830871582032,
|
| 3348 |
+
"mean_token_accuracy": 0.6615023009479046,
|
| 3349 |
+
"num_tokens": 19846593.0,
|
| 3350 |
+
"step": 3340
|
| 3351 |
+
},
|
| 3352 |
+
{
|
| 3353 |
+
"entropy": 1.7411245226860046,
|
| 3354 |
+
"epoch": 5.756501182033097,
|
| 3355 |
+
"grad_norm": 0.8770884871482849,
|
| 3356 |
+
"learning_rate": 6.604e-05,
|
| 3357 |
+
"loss": 1.8150835037231445,
|
| 3358 |
+
"mean_token_accuracy": 0.6629907101392746,
|
| 3359 |
+
"num_tokens": 19901728.0,
|
| 3360 |
+
"step": 3350
|
| 3361 |
+
},
|
| 3362 |
+
{
|
| 3363 |
+
"entropy": 1.7868980005383492,
|
| 3364 |
+
"epoch": 5.773694390715667,
|
| 3365 |
+
"grad_norm": 0.8012292385101318,
|
| 3366 |
+
"learning_rate": 6.564e-05,
|
| 3367 |
+
"loss": 1.8411848068237304,
|
| 3368 |
+
"mean_token_accuracy": 0.6610983822494745,
|
| 3369 |
+
"num_tokens": 19962294.0,
|
| 3370 |
+
"step": 3360
|
| 3371 |
+
},
|
| 3372 |
+
{
|
| 3373 |
+
"entropy": 1.8151665195822715,
|
| 3374 |
+
"epoch": 5.790887599398237,
|
| 3375 |
+
"grad_norm": 0.6628616452217102,
|
| 3376 |
+
"learning_rate": 6.524e-05,
|
| 3377 |
+
"loss": 1.873934555053711,
|
| 3378 |
+
"mean_token_accuracy": 0.6563344091176987,
|
| 3379 |
+
"num_tokens": 20023732.0,
|
| 3380 |
+
"step": 3370
|
| 3381 |
+
},
|
| 3382 |
+
{
|
| 3383 |
+
"entropy": 1.7544716522097588,
|
| 3384 |
+
"epoch": 5.808080808080808,
|
| 3385 |
+
"grad_norm": 0.765192985534668,
|
| 3386 |
+
"learning_rate": 6.484e-05,
|
| 3387 |
+
"loss": 1.7676244735717774,
|
| 3388 |
+
"mean_token_accuracy": 0.6681830242276192,
|
| 3389 |
+
"num_tokens": 20084955.0,
|
| 3390 |
+
"step": 3380
|
| 3391 |
+
},
|
| 3392 |
+
{
|
| 3393 |
+
"entropy": 1.7742430947721004,
|
| 3394 |
+
"epoch": 5.8252740167633785,
|
| 3395 |
+
"grad_norm": 0.9363911747932434,
|
| 3396 |
+
"learning_rate": 6.444e-05,
|
| 3397 |
+
"loss": 1.7936756134033203,
|
| 3398 |
+
"mean_token_accuracy": 0.6640898622572422,
|
| 3399 |
+
"num_tokens": 20145523.0,
|
| 3400 |
+
"step": 3390
|
| 3401 |
+
},
|
| 3402 |
+
{
|
| 3403 |
+
"entropy": 1.7672609627246856,
|
| 3404 |
+
"epoch": 5.842467225445949,
|
| 3405 |
+
"grad_norm": 0.8024185299873352,
|
| 3406 |
+
"learning_rate": 6.404e-05,
|
| 3407 |
+
"loss": 1.8390132904052734,
|
| 3408 |
+
"mean_token_accuracy": 0.66049126945436,
|
| 3409 |
+
"num_tokens": 20207613.0,
|
| 3410 |
+
"step": 3400
|
| 3411 |
+
},
|
| 3412 |
+
{
|
| 3413 |
+
"entropy": 1.7419263988733291,
|
| 3414 |
+
"epoch": 5.85966043412852,
|
| 3415 |
+
"grad_norm": 0.7452662587165833,
|
| 3416 |
+
"learning_rate": 6.364e-05,
|
| 3417 |
+
"loss": 1.7592693328857423,
|
| 3418 |
+
"mean_token_accuracy": 0.6729626737534999,
|
| 3419 |
+
"num_tokens": 20264080.0,
|
| 3420 |
+
"step": 3410
|
| 3421 |
+
},
|
| 3422 |
+
{
|
| 3423 |
+
"entropy": 1.794348457455635,
|
| 3424 |
+
"epoch": 5.876853642811089,
|
| 3425 |
+
"grad_norm": 0.7698886394500732,
|
| 3426 |
+
"learning_rate": 6.324e-05,
|
| 3427 |
+
"loss": 1.8215929031372071,
|
| 3428 |
+
"mean_token_accuracy": 0.6594825953245163,
|
| 3429 |
+
"num_tokens": 20325935.0,
|
| 3430 |
+
"step": 3420
|
| 3431 |
+
},
|
| 3432 |
+
{
|
| 3433 |
+
"entropy": 1.8058848246932029,
|
| 3434 |
+
"epoch": 5.89404685149366,
|
| 3435 |
+
"grad_norm": 0.7813654541969299,
|
| 3436 |
+
"learning_rate": 6.284e-05,
|
| 3437 |
+
"loss": 1.8705635070800781,
|
| 3438 |
+
"mean_token_accuracy": 0.6575064200907945,
|
| 3439 |
+
"num_tokens": 20384219.0,
|
| 3440 |
+
"step": 3430
|
| 3441 |
+
},
|
| 3442 |
+
{
|
| 3443 |
+
"entropy": 1.8296722590923309,
|
| 3444 |
+
"epoch": 5.9112400601762305,
|
| 3445 |
+
"grad_norm": 0.7985308766365051,
|
| 3446 |
+
"learning_rate": 6.244e-05,
|
| 3447 |
+
"loss": 1.8622390747070312,
|
| 3448 |
+
"mean_token_accuracy": 0.6598837457597255,
|
| 3449 |
+
"num_tokens": 20442055.0,
|
| 3450 |
+
"step": 3440
|
| 3451 |
+
},
|
| 3452 |
+
{
|
| 3453 |
+
"entropy": 1.745755286514759,
|
| 3454 |
+
"epoch": 5.928433268858801,
|
| 3455 |
+
"grad_norm": 0.7957124710083008,
|
| 3456 |
+
"learning_rate": 6.204e-05,
|
| 3457 |
+
"loss": 1.778817367553711,
|
| 3458 |
+
"mean_token_accuracy": 0.6711658544838428,
|
| 3459 |
+
"num_tokens": 20500787.0,
|
| 3460 |
+
"step": 3450
|
| 3461 |
+
},
|
| 3462 |
+
{
|
| 3463 |
+
"entropy": 1.8037077650427817,
|
| 3464 |
+
"epoch": 5.945626477541371,
|
| 3465 |
+
"grad_norm": 0.8097943067550659,
|
| 3466 |
+
"learning_rate": 6.164e-05,
|
| 3467 |
+
"loss": 1.8763154983520507,
|
| 3468 |
+
"mean_token_accuracy": 0.6581781908869744,
|
| 3469 |
+
"num_tokens": 20560544.0,
|
| 3470 |
+
"step": 3460
|
| 3471 |
+
},
|
| 3472 |
+
{
|
| 3473 |
+
"entropy": 1.6478220209479333,
|
| 3474 |
+
"epoch": 5.962819686223941,
|
| 3475 |
+
"grad_norm": 0.7882372736930847,
|
| 3476 |
+
"learning_rate": 6.124e-05,
|
| 3477 |
+
"loss": 1.7081596374511718,
|
| 3478 |
+
"mean_token_accuracy": 0.6812212504446507,
|
| 3479 |
+
"num_tokens": 20621866.0,
|
| 3480 |
+
"step": 3470
|
| 3481 |
+
},
|
| 3482 |
+
{
|
| 3483 |
+
"entropy": 1.7625097312033176,
|
| 3484 |
+
"epoch": 5.980012894906512,
|
| 3485 |
+
"grad_norm": 0.780114471912384,
|
| 3486 |
+
"learning_rate": 6.084000000000001e-05,
|
| 3487 |
+
"loss": 1.7864303588867188,
|
| 3488 |
+
"mean_token_accuracy": 0.6720023825764656,
|
| 3489 |
+
"num_tokens": 20677413.0,
|
| 3490 |
+
"step": 3480
|
| 3491 |
+
},
|
| 3492 |
+
{
|
| 3493 |
+
"entropy": 1.7418652877211571,
|
| 3494 |
+
"epoch": 5.9972061035890825,
|
| 3495 |
+
"grad_norm": 0.8374961614608765,
|
| 3496 |
+
"learning_rate": 6.044000000000001e-05,
|
| 3497 |
+
"loss": 1.8426128387451173,
|
| 3498 |
+
"mean_token_accuracy": 0.6612365163862706,
|
| 3499 |
+
"num_tokens": 20737424.0,
|
| 3500 |
+
"step": 3490
|
| 3501 |
+
},
|
| 3502 |
+
{
|
| 3503 |
+
"entropy": 1.8112690711950328,
|
| 3504 |
+
"epoch": 6.013754566946056,
|
| 3505 |
+
"grad_norm": 0.7742412686347961,
|
| 3506 |
+
"learning_rate": 6.004000000000001e-05,
|
| 3507 |
+
"loss": 1.8320732116699219,
|
| 3508 |
+
"mean_token_accuracy": 0.6643109286760355,
|
| 3509 |
+
"num_tokens": 20795175.0,
|
| 3510 |
+
"step": 3500
|
| 3511 |
}
|
| 3512 |
],
|
| 3513 |
"logging_steps": 10,
|
|
|
|
| 3527 |
"attributes": {}
|
| 3528 |
}
|
| 3529 |
},
|
| 3530 |
+
"total_flos": 1.7073275440778445e+17,
|
| 3531 |
"train_batch_size": 2,
|
| 3532 |
"trial_name": null,
|
| 3533 |
"trial_params": null
|